Key words: ESBLs; Ceftriaxone; Escherichia coli; MICs; Bacteremia

Transcription

1 JCM Accepts, published online ahead of print on 16 April 2014 J. Clin. Microbiol. doi: /jcm Copyright 2014, American Society for Microbiology. All Rights Reserved Title: Determining the Optimal Ceftriaxone MIC to Trigger ESBL Confirmatory Testing Running Title: ESBL Testing Authors and Affiliations: Yanjie Huang 1, Karen C. Carroll 2, Sara E. Cosgrove 3, & Pranita D. Tamma# 4, M.D., M.H.S. 1 Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health 2 Department of Pathology, Division of Medical Microbiology, Johns Hopkins Medical Institutions 3 Department of Medicine, Division of Infectious Diseases, Johns Hopkins Medical Institutions 4 Department of Pediatrics, Division of Infectious Diseases, Johns Hopkins Medical Institutions Pranita D. Tamma, MD, MHS (#Corresponding author) Johns Hopkins Medical Institutions Department of Pediatrics, Division of Infectious Diseases 200 North Wolfe Street, Suite 3155 Baltimore, Maryland Tel (443) Fax (410) ptamma1@jhmi.edu Key words: ESBLs; Ceftriaxone; Escherichia coli; MICs; Bacteremia 1

2 Abstract As routine testing of clinical isolates for ESBL production (screen plus phenotypic confirmatory testing) is no longer required, a number of clinical microbiology laboratories use ceftriaxone minimum inhibitory concentrations (MICs) as proxies for identifying bacteria as potential ESBL producers. Data from our institution suggest that a ceftriaxone MIC cutoff of 8 µg/ml is an excellent predictor of ESBL production with a positive predictive value and negative predictive value approaching 100% and 99.5%, respectively. Manuscript In June 2010, the Clinical Laboratory and Standards Institute (CLSI) lowered the ceftriaxone (and cefotaxime) breakpoint from 8 µg/ml to 1 µg/ml, and with that change, the recommendation for screening for extended-spectrum β-lactamase (ESBL) production became optional 1. This was in part due to the large workload imposed on clinical microbiology laboratories with phenotypic confirmatory ESBL testing 1. However, identification of organisms producing ESBLs still has important clinical implications. A portion of these organisms retain in vitro susceptibility to piperacillin-tazobactam and cefepime 2, even though these are generally considered inferior agents to carbapenem therapy for the treatment of invasive infections by ESBL- producing organisms 3-5. Without a method of alerting clinicians to the possibility that an agent may be an ESBL producer, potentially suboptimal therapy (e.g., cefepime or piperacillin-tazobactam) may be inadvertently prescribed. In addition, the absence of institutional epidemiological data 2

3 on ESBLs can hamper efforts to control their spread within healthcare facilities 6. For these reasons, many clinical microbiology laboratories use ceftriaxone (or other thirdgeneration cephalosporin) MIC thresholds to indicate to clinicians that an organism is a possible ESBL producer or to trigger additional confirmatory phenotypic testing. If the ceftriaxone MIC threshold is lower than necessary, this practice has the potential to lead to the overuse of carbapenems or to needlessly increase the workload of microbiology laboratories (for institutions where confirmatory testing is still performed). If the ceftriaxone MIC threshold for ESBL identification is too high, appropriate infection control and treatment strategies may not be deployed. Our objective was to evaluate the sensitivity and specificity of various ceftriaxone MICs in determining the optimal threshold for identifying an organism as a potential ESBL producer. Microbiology methods. Blood isolates growing Escherichia coli, Klebsiella pneumoniae, Klebsiella oxytoca, and Proteus mirabilis from January 2007-December 2013 were included. These organisms were selected as the CLSI recommended method for initial screening and phenotypic confirmatory testing is limited to these organisms 7. Clinical samples were processed at the Johns Hopkins Hospital Microbiology Laboratory according to standard operating procedures. Antimicrobial susceptibility testing was determined by the BD Phoenix Automated System (BD Diagnostics, Sparks, MD). Klebsiella pneumoniae, Klebsiella oxytoca, Escherichia coli, and Proteus mirabilis organisms with MICs 2 µg/ml for ceftriaxone underwent further screening for ESBL production. A decrease of >3 doubling dilutions in the MIC for 3

4 either ceftriaxone or ceftazidime tested in combination with 4 µg/ml of clavulanic acid, versus its MIC when tested alone, was used to confirm ESBL status Statistical Approach: Statistical analyses were performed using the statistical R package (3.0.2) (R Core Team, Vienna, Austria). The Chi-square test was used to analyze categorical variables. A two-tailed p-value of <0.05 was considered statistically significant. A Receiver operating characteristic (ROC) curve was used to determine the optimal breakpoint for the detection of ESBL producing organisms using various ceftriaxone MICs. The discriminatory power was evaluated by the area under the ROC curve (AUC) with an AUC value of 0.5 indicating no discriminative ability and an AUC exceeding 0.8 indicating good to excellent prediction. The sensitivity and specificity of the prediction rule were calculated at various ceftriaxone MIC values. During the seven year study period, there were 1386 unique episodes of bacteremia with Escherichia coli, Klebsiella species, or Proteus mirabilis at The Johns Hopkins Hospital. After 18 cases of carbapenem-resistant Enterobacteriaceae were excluded, there were 270 and 1116 cases of ESBL and non-esbl producing bacteremia, respectively. The proportion of ESBL producing organisms was 12.9% in 2007 and 21.9% in 2013 (p < 0.01), with a gradual increase observed annually. Evaluating all bloodstream isolates, 20.0% of E. coli, 19.1% of Klebsiella spp., and 19.1% of P. mirabilis were ESBL producers. The median MIC of ceftriaxone among ESBL producing organisms was 64µg/ml. The distribution of MICs by ESBL and non-esbl producing organisms is depicted in Figure 1. 4

5 We determined the sensitivity and specificity of detecting ESBL producing organisms using various MICs of ceftriaxone by developing an ROC curve. The AUC of the ROC curve was 0.99 (Figure 2). An MIC cutoff of 8 µg/ml had the greatest overall sensitivity, specificity, positive predictive value, and negative predictive value at 97.8%, 100%, 100%, and 99.5%, respectively. There were 5 (1.5%) ESBL producing organisms in our cohort with ceftriaxone MICs <8 µg/ml. With a ceftriaxone MIC threshold of 2 µg/ml (as recommended by the CDC to consider an organism as a potential ESBL-producer 7 ), the sensitivity, specificity, positive predictive value, and negative predictive value would be 98.1%, 91.6%, 73.8%, and 99.5%, respectively. Our data suggest that a ceftriaxone MIC cutoff of 8 µg/ml against E. coli, Klebsiella spp., and P. mirabilis is an excellent predictor of ESBL production with a positive predictive value and negative predictive value approaching 100% and 99.5%, respectively. With lower thresholds, even though the sensitivity of detecting ESBLs would be relatively unchanged (i.e., the likelihood of detecting additional ESBL producers would be low), the specificity of detecting ESBLs would be compromised. A decreased specificity or overcalling organisms as potential ESBL producers could mean a subsequent increase in carbapenem use, leading to an additional strain on our already constrained antibiotic armamentarium 8,9. Identification of ESBL producing organisms has important implications. Because they can spread rapidly between patients in healthcare institutions 6,10, their identification 5

6 warrants the implementation of appropriate infection control measures. Additionally, as they may appear susceptible in vitro to cefepime and piperacilin-tazobactam 2, even though some experts believe these agents to be inferior to carbapenems for invasive ESBL infections 3,4, their recognition can impact antibiotic treatment decisions Certain Enterobacteriaceae may express either plasmid-mediated or chromosomallymediated AmpC β-lactamases 11. Using the current CLSI ceftriaxone breakpoint of 1 µg/ml, we previously found that 100% of 96 organisms expressing AmpC β-lactamases had ceftriaxone MICs 1 µg/ml 11. Ceftriaxone has been found to be problematic for the treatment of these organisms but the same has not been demonstrated with broaderspectrum β-lactams 5. Therefore, co-existence of AmpC β-lactamases should be unlikely to impact the ceftriaxone MIC used to trigger ESBL testing. As our study consists of single center data, our range of MICs for ESBL producing organisms could differ from those of other centers. However, we still believe this study is important in reminding us that clinical microbiology laboratories should periodically review their institutional data to determine the most accurate ceftriaxone MIC threshold to identify an organism as potentially ESBL producing. This is true whether laboratories use ceftriaxone MICs as a proxy for possible ESBL production or whether ceftriaxone MICs are used to trigger further phenotypic testing. If ceftriaxone MIC thresholds indicating potential ESBL production are unnecessarily low, the former could lead to the overuse of carbapenems when other β-lactams could suffice and the latter could lead to 6

7 needless additional testing steps by already overburdened clinical microbiology laboratories We realize that our findings are preliminary and we recommend that other clinical microbiology laboratories repeat our study using their institutional data and consider expanding it to other Enterobacteriaceae. If our results are replicated, then Enterobacteriaceae with ceftriaxone MICs of 8 μg/ml or less should be considered as having a low likelihood of ESBL production, and treatment should be guided by the individual organism antibiogram (i.e., piperacillin-tazobactam and cefepime can be considered if the isolate is susceptible). Acknowledgements This work was supported by a Thrasher Research Foundation award to P.D.T. S.E.C and P.D.T. receive salary support from a Pfizer Independent Grant for Learning and Change, unrelated to the current project. References 1. Dudley MN, Ambrose PG, Bhavnani SM, Craig WA, Ferraro MJ, Jones RN. Background and rationale for revised clinical and laboratory standards institute interpretive criteria (Breakpoints) for Enterobacteriaceae and Pseudomonas aeruginosa: I. Cephalosporins and Aztreonam Clin Infect Dis 56:

8 Marchaim D, Sunkara B, Lephart PR, Gudur UM, Bhargava A, Mynatt RP, Zhao JJ, Bheemreddy S, Hayakawa K, Chopra T, Dhar S, Kaye KS Extended- Spectrum beta-lactamase Producers Reported as Susceptible to Piperacillin- Tazobactam, Cefepime, and Cefuroxime in the Era of Lowered Breakpoints and No Confirmatory Tests. Infect Control Hosp Epidemiol 33: Perez F, Bonomo RA. Can we really use ss-lactam/ss-lactam inhibitor combinations for the treatment of infections caused by extended-spectrum sslactamase-producing bacteria? Clin Infect Dis 54: Lee NY, Lee CC, Huang WH, Tsui KC, Hsueh PR, Ko WC. Cefepime Therapy for Monomicrobial Bacteremia Caused by Cefepime-Susceptible Extended- Spectrum Beta-Lactamase-Producing Enterobacteriaceae: MIC Matters Clin Infect Dis 56: Hsu AJ, Tamma PD The Treatment of Multi-Drug Resistant Gramnegative Infections in Children. Clininfect dis [Epub ahead of print] 6. Oteo J, Navarro C, Cercenado E, Delgado-Iribarren A, Wilhelmi I, Orden B, Garcia C, Miquelanez S, Perez-Vazquez, Garcia-Cobos S, Aracil B, Bautista V, Campos J. Spread of Escherichia coli strains with high-level cefotaxime and ceftazidime resistance between the community, long-term care facilities, and hospital institutions J Clin Microbiol 44: Wayne PA. Performance standards for antimicrobial susceptibility testing; 18th informational supplement Clin Lab Standards Inst. M100 S Tamma PD, Wu H, Gerber JS, Hsu AJ, Tekle T, Carroll KC, Cosgrove SE Outcomes of children with Enterobacteriaceae bacteremia with reduced 8

9 susceptibility to ceftriaxone: do the revised breakpoints translate to improved patient outcomes? Pediatr Infect Dis J 32: Tamma PD, Powers JH Do patient data really support the clinical and laboratory standards institute recomendation for lowering third-generation cephalosporin interpretve breakpoints? Clin Infect Dis 57: Tamma PD, Savard P, Pal T, Sonnevend A, Perl TM, Milstone AM An outbreak of extended-spectrum beta-lactamase-producing Klebsiella pneumoniae in a neonatal intensive care unit. Infect Control Hosp Epidemiol 33: Tamma PD, Girdwood SCT, Gopaul R, Tekle T, Roberts AA, Harris AD, Carroll KC Clin Infect Dis 57:

10 Downloaded from Figure 1: The distribution of minimum inhibitory concentrations (MICs) by extendedspectrum β-lactamase (ESBL) producing organisms and non-esbl producing organisms, excluding carbapenem-resistant Enterobacteriaceae on July 5, 2018 by guest

11 Downloaded from Figure 2: Receiver-operator characteristic curves for ceftriaxone minimum inhibitory concentrations for the detection of extended-spectrum β-lactamase producing Enterobacteriaceae on July 5, 2018 by guest